ML20072A823

Ro:On 821204,power Calorimetric Performed to Determine Agreement Between Indicated Power & Calculated Actual Power Based on Pool Heatup Rate.Results Showed Minus 30.7% Error. on 821206,calorimetric Showed Plus 56.6% Error

ML20072A823
Person / Time
Site:	05000128
Issue date:	12/22/1982
From:	Rogers D TEXAS A&M UNIV., COLLEGE STATION, TX
To:	Madsen G NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION IV)
References
NUDOCS 8301240143
Download: ML20072A823 (3)

Text

• .

.*q , .P TEXAS ENGINEERING EXPERI AIENT STATION THE TEXAS A&M UNIVERSITY SYSTEM COLLEGE STATION, TEXAS 77843

('

M J 22 December 1982 NUCLEAR SCIENOE CENTER 713/845-7551 Mr. G. L. Madsen Office of Inspection and Enforcement Region IV U.S. Nuclear Regulatory Commission -

611 Ryan Plaza, Suite 1000 Arlington, Texas 76012

Subject:

Reactor Operation in Excess of Licensed Power Level Reported 12-17-82

Dear Mr. Madsen:

Enclosed is a final report of a reportable occurrence observed during reactor operations of the NSCR. This report is submitted in compliance with Section 6.4 and 1.8 of Change No. 11 of the Technical Specifications, Facility License No. R-83 for the Nuclear Science Center, Texas A&M University Sincerely, Dale Rogers,i tager Reactor Operations DR/nm Enclosure il

/

8301240143 B21222 PDR ADOCK 05000128 ,

S PDR _

RESEARCH AND DEVELOPMENT FOR MANKIND

Final Report of Violation Due to Reactor Operation in Excess of Licensed Power Level R_eportabl_e Occurrence _

On December 4, 1982 following a core change to establish Core VIII a power calorimetric was perfonned to determine the agreement between indicated power and calculated actual power based on the pool heatup rate. The results of this calorimetric showed a -30.7% error (indicated power 400 Kw, calculated power 277 Kw). The data was reviewed by the Manager of Reactor Operations, and the large discrepancy was questioned. However, a review of all procedural regirements and initial conditions revealed no deviations. The reactor power detectors were then adjusted to agree with the results obtained. Reactor power was increased to 1 Mw in 100 Kw increments to determine the power coefficient, and the reactor was then shutdown for the remainder of the weekend.

On Monday, December 6, 1982 the reactor was operated at 1 Mw (indicated) and reactor operations personnel noted that fuel temperature, radiction levels on the reactor bridge, and contral rod positions were all somewhat higher than expected.

In discussing this with the Acting Director, the Mar.ager of Reactor Operations had the operations personnel operate at a reduced power level (90% indicated) until

a second calorimetric could be performed. This was completed on December 10, 1982 and showed a +56.6% error (indicated power 400 Kw, calculated power 626 Kw). This large discrepancy in the other direction could not be explained and made the calorimetric procedure suspect. At this time it was felt that positioning of the reactor within the stall might contribute to errors in the results since no proce-dural deviations could be found. Based on these results, however, the detectors were again repositioned and, reactor operation continued until another calorimetric could be completed on December 13, 1982. This calorimetric indicated a -7.6% error.

Two additional calorimetrics run on December 16, 1982 indicated -1,65% error and

-12.4% error.

Based on a review of the data obtained in the series of calorimetrics it was decided by NSC management on December 17, 1982 that the reactor could possibly have operated in excess of its licensed power level of IMw. NRC representative G.L.

Constable was contacted by phone to discuss the situation. In addition, NRC Region IV was contacted on that same date.

On Monday, December 20, 1982 Mr. Constable reviewed calorimetric data and discussed the sequence of events with reactor operations personnel. In addition, a calorime-tric was performed under his supervision, and the results indicated a -7.3% error.

Since the last four calorimetrics indicated that the results of 12-10-82 may have overestimated power it is felt that actual reactor power may have been as high as 1.37 Mw on December 6, 1982 and 1.23 Mw on December 7 to December 9, 1982.

Corr _ectiv_e Action _

The series of calorimetrics performed with the reactor in various locations indi-cated that positioning of the reactor within the stall probably was not the source

,. g .,

of error in the calorimetric performed on December 4, 1982. During this period, however, the ice bath used as a reference junction for the thermocouples was looked at very closely as being a possible source of error. Various combinations of shorting thermocouples were trie<j, and variance of reference temperature was also considered. A test performed on December 21, 1982 demonstrated that a crushed ice bath composed primagily of ice g with very little water increased in tenperature from approximately 31.2 F to 31.8 F over a two hour period. It was determined that this minor change in reference temperature could contribute as much as 35%

error to calorimetric results. In discussing the preparation of the ice bath with personnel involved it is felt that this could quite possibly have been the cause of the excessive deviation on December 4. As a corrective action for future calorimetrics proper training will be conducted with all operations personnel to demonstrate preparation of a proper ice bath slurry. In addition, a thermometer will be used to check ice bath temperature prior to each reading to insure a constant reference tenperature is being maintained. Also, the thermocouple halder will be modified to insure that no thermocouples will be s.iorted,to each other or to the stainless steel insulated can.

.